Keypoint Transfer Segmentation

Keypoint Transfer for Fast Whole-Body Segmentation

We introduce an approach for image segmentation based on sparse correspondences between keypoints in testing and training images. Keypoints represent automatically identified distinctive image locations, where each keypoint correspondence suggests a transformation between images. We use these correspondences to transfer the label maps of entire organs from the training images to the test image. The keypoint transfer algorithm includes three steps: 1) keypoint matching; 2) voting-based keypoint labeling; and 3) keypoint-based probabilistic transfer of organ segmentations. We report segmentation results for abdominal organs in whole-body CT and MRI, as well as in contrast-enhanced CT and MRI. Our method offers a speed-up of about three orders of magnitude in comparison with common multi-atlas segmentation while achieving an accuracy that compares favorably. Moreover, keypoint transfer does not require the registration to an atlas or a training phase. Finally, the method allows for the segmentation of scans with a highly variable field-of-view.

Delineation of a Cardiac Planning Organ-At-Risk Volume Using Real-Time Magnetic Resonance Imaging for Cardiac Protection in Thoracic and Breast Radiation Therapy

PURPOSE

Cardiac radiation is associated with cardiotoxicity in patients with thoracic and breast malignancies. We conducted a prospective study using cine magnetic resonance imaging (MRI) scans to evaluate heart motion. We hypothesized that cine MRI could be used to define population-based cardiac planning organ-at-risk volumes (PRV).

METHODS AND MATERIALS

A total of 16 real-time acquisitions were obtained per subject on a 1.5 Tesla MRI (Philips Ingenia). Planar cine MRI was performed in 4 sequential sagittal and coronal planes at free-breathing (FB) and deep-inspiratory breath hold (DIBH). In-plane cardiac motion was assessed using a scale-invariant feature transformation-based algorithm. Subject-specific pixel motion ranges were defined in anteroposterior (AP), left-right (LR), and superoinferior (SI) planes. Averages of the 98% and 67% of the maximum ranges of pixel displacement were defined by subject, then averaged across the cohort to calculate PRV expansions at FB and DIBH.

RESULTS

Data from 20 subjects with a total of 3120 image frames collected per subject in coronal and sagittal planes at DIBH and FB, and 62,400 total frames were analyzed. Cohort averages of 98% of the maximum cardiac motion ranges comprised margin expansions of 12.5 ± 1.1 mm SI, 5.8 ± 1.2 mm AP, and 6.6 ± 1.0 mm LR at FB and 6.7 ± 1.5 mm SI, 4.7 ± 1.3 mm AP, and 5.3 ± 1.3 mm LR at DIBH. Margins for 67% of the maximum range comprised 7.7 ± 0.7 mm SI, 3.2 ± 0.6 mm AP, and 3.7 ± 0.6 mm LR at FB and 4.1 ± 0.9 mm SI, 2.7 ± 0.8 mm AP, and 3.2 ± 0.8 mm LR at DIBH. Subsequently, these margins were simplified to form PRVs for treatment planning.

CONCLUSIONS

We implemented scale-invariant feature transformation-based motion tracking for analysis of the cardiac cine MRI scans to quantify motion and create cohort-based cardiac PRVs to improve cardioprotection in breast and thoracic radiation.

Automatic Multiorgan Segmentation via Multiscale Registration and Graph Cut

We propose an automatic multiorgan segmentation method for 3-D radiological images of different anatomical contents and modalities. The approach is based on a simultaneous multilabel graph cut optimization of location, appearance, and spatial configuration criteria of target structures. Organ location is defined by target-specific probabilistic atlases (PA) constructed from a training dataset using a fast (2+1)D SURF-based multiscale registration method involving a simple four-parameter transformation. PAs are also used to derive target-specific organ appearance models represented as intensity histograms. The spatial configuration prior is derived from shortest-path constraints defined on the adjacency graph of structures. Thorough evaluations on Visceral project benchmarks and training dataset, as well as comparisons with the state-of-the-art confirm that our approach is comparable to and often outperforms similar approaches in multiorgan segmentation, thus proving that the combination of multiple suboptimal but complementary information sources can yield very good performance.

Stratified Decision Forests for Accurate Anatomical Landmark Localization in Cardiac Images

Accurate localization of anatomical landmarks is an important step in medical imaging, as it provides useful prior information for subsequent image analysis and acquisition methods. It is particularly useful for initialization of automatic image analysis tools (e.g. segmentation and registration) and detection of scan planes for automated image acquisition. Landmark localization has been commonly performed using learning based approaches, such as classifier and/or regressor models. However, trained models may not generalize well in heterogeneous datasets when the images contain large differences due to size, pose and shape variations of organs. To learn more data-adaptive and patient specific models, we propose a novel stratification based training model, and demonstrate its use in a decision forest. The proposed approach does not require any additional training information compared to the standard model training procedure and can be easily integrated into any decision tree framework. The proposed method is evaluated on 1080 3D high-resolution and 90 multi-stack 2D cardiac cine MR images. The experiments show that the proposed method achieves state-of-the-art landmark localization accuracy and outperforms standard regression and classification based approaches. Additionally, the proposed method is used in a multi-atlas segmentation to create a fully automatic segmentation pipeline, and the results show that it achieves state-of-the-art segmentation accuracy.